Legal claims defining the scope of protection. Each claim is shown in both the original legal language and a plain English translation.
1. A system for predicting movements, the system comprising: one or more processors and a memory, the memory being a non-transitory computer-readable medium having executable instructions encoded thereon, such that upon execution of the instructions, the one or more processors perform operations of: computing relative positions of multiple objects of interest; generating a feature representation by forming a matrix based on the relative positions; predicting movement of the multiple objects of interest by applying clustering to the feature representation and by performing canonical correlation analysis; and controlling a device based on the predicted movement of the multiple objects of interest.
The system predicts movements of multiple objects of interest by analyzing their relative positions and controlling devices based on the predictions. The system uses one or more processors and a non-transitory memory storing executable instructions. The instructions cause the processors to compute the relative positions of the objects, then generate a feature representation by forming a matrix from these positions. The system predicts movement by applying clustering to the feature representation and performing canonical correlation analysis. The predictions are used to control a device, such as a robot, autonomous vehicle, or surveillance system. The clustering groups similar movement patterns, while canonical correlation analysis identifies relationships between the objects' movements and external factors. This approach improves accuracy in predicting interactions between objects, enabling proactive device control in dynamic environments. The system is applicable in autonomous navigation, collision avoidance, and human-robot interaction scenarios.
2. The system of claim 1 , wherein the device includes a display.
A system for managing data processing tasks includes a device with a display. The device is configured to receive input data, process the data according to predefined rules, and generate output data. The display is used to present the processed data or related information to a user. The system may also include additional components such as sensors, communication modules, or storage units to enhance functionality. The device processes data by executing algorithms or logic stored in memory, ensuring accurate and efficient task execution. The display provides visual feedback, allowing users to monitor operations, review results, or interact with the system. The system may be used in various applications, including industrial automation, data analysis, or user interface-driven workflows, where real-time data processing and visualization are essential. The inclusion of a display enables user interaction and improves usability by presenting processed data in a structured format. The system ensures reliable data handling while maintaining flexibility for different processing requirements.
3. The system of claim 1 , wherein the device includes a motor.
A system for controlling a device includes a motor integrated into the device to provide mechanical movement or actuation. The motor is configured to drive one or more components of the device, enabling precise control over its operation. The system may include sensors to monitor the device's state, such as position, speed, or force, and a controller that processes sensor data to adjust motor operation in real time. The controller can implement feedback loops to maintain desired performance parameters, such as maintaining a target speed or position despite external disturbances. The motor may be an electric motor, such as a brushless DC motor, stepper motor, or servo motor, depending on the application requirements. The system can be used in various applications, including robotics, automation, industrial machinery, or consumer electronics, where controlled movement is essential. The motor's integration allows for compact, efficient, and responsive device operation, improving overall system performance and reliability.
4. The system of claim 1 , wherein the one or more processors further perform the operation of generating pairs of tactical feature vectors.
The system relates to a data processing system for analyzing tactical data, such as military or security operations, to improve decision-making. The problem addressed is the need to extract meaningful patterns from complex tactical data to enhance situational awareness and response strategies. The system includes one or more processors configured to process tactical data, such as sensor inputs, communication logs, or operational reports, to identify relevant features. These features are then used to generate tactical feature vectors, which are numerical representations of key attributes from the data. The system further generates pairs of these tactical feature vectors to enable comparative analysis, such as identifying similarities, differences, or correlations between different tactical scenarios. This pairing allows for more nuanced insights, such as detecting emerging threats, optimizing resource allocation, or refining response protocols. The system may also include a user interface to display the analyzed data and feature vectors, aiding human operators in interpreting the results. The overall goal is to provide a structured and automated way to derive actionable intelligence from raw tactical data, improving efficiency and accuracy in decision-making processes.
5. The system of claim 4 , wherein the canonical correlation analysis is performed using the pairs of tactical feature vectors.
This invention relates to a system for analyzing tactical data using canonical correlation analysis (CCA). The system addresses the challenge of extracting meaningful relationships between different sets of tactical features in military or defense applications, where data from sensors, communications, or other sources must be correlated to improve situational awareness or decision-making. The system includes a data processing module that receives tactical feature vectors from multiple sources. These vectors represent different aspects of a tactical scenario, such as sensor readings, communication patterns, or environmental conditions. The system performs canonical correlation analysis on pairs of these tactical feature vectors to identify linear relationships between them. This analysis helps uncover hidden dependencies that may not be apparent through traditional statistical methods. The CCA process involves computing canonical variates, which are linear combinations of the original feature vectors that maximize their correlation. By analyzing these variates, the system can determine how different tactical features interact, enabling more accurate predictions or classifications in real-time scenarios. The results of the CCA are used to enhance decision support systems, improve threat detection, or optimize resource allocation in dynamic environments. The system is particularly useful in military applications where rapid and accurate correlation of diverse data streams is critical. By leveraging CCA, it provides a robust framework for extracting actionable insights from complex tactical data.
6. The system as set forth in claim 1 , wherein controlling the device includes causing a camera to orient based on the predicted movement.
A system for controlling a camera device based on predicted movement. The system operates in the domain of automated camera orientation, addressing the problem of maintaining optimal camera alignment with a moving subject or scene. The system predicts the movement of a subject or scene and adjusts the camera's orientation accordingly to track or follow the movement. This ensures continuous and accurate capture of the subject or scene, even as it moves. The system may use sensors, algorithms, or other input data to predict movement and then control the camera's orientation mechanisms, such as motors or actuators, to adjust its position. The system may also integrate with other components, such as image processing modules or user interfaces, to enhance tracking performance or provide additional functionality. The camera may be part of a larger surveillance, security, or imaging system, where maintaining alignment with a moving target is critical. The system improves upon traditional manual or fixed-position cameras by automating the orientation process, reducing the need for human intervention, and increasing the reliability of the captured footage.
7. The system as set forth in claim 1 , wherein the canonical correlation analysis (CCA) maximizes the following objective function: CCA comp = arg max u , w ∑ n = 1 N ( u T h n ) ( v n T w ) ∑ n = 1 N u T h n h n T u ∑ n = 1 N w T v n v n T w = arg max u , w u T C hv w u T C hh u w T C vv w wherein u and w are CCA components that project data onto a shared embedding and C hh , C vv , C hv are covariance matrices, the tactical formations of a home team and an adversary team are embedded into vectors h and v, respectively, N is the total number of tactical formations during a given time period, and the multiple objects of interest are the members of the adversary team.
This invention relates to a system for analyzing tactical formations in military or competitive scenarios using canonical correlation analysis (CCA). The system addresses the challenge of understanding and predicting the movements and strategies of opposing teams by embedding their formations into vector representations and identifying correlations between them. The CCA component of the system maximizes an objective function that aligns the projections of these vectors, optimizing the relationship between the formations of a home team and an adversary team. The objective function involves covariance matrices derived from the embedded vectors, which capture the statistical dependencies between the formations. By projecting the data onto a shared embedding space, the system enables the identification of patterns and correlations that can inform strategic decision-making. The method processes multiple tactical formations over a given time period, with the adversary team members serving as the objects of interest. This approach enhances situational awareness and predictive capabilities in dynamic, adversarial environments.
8. A computer program product for predicting movements, the computer program product comprising: a non-transitory computer-readable medium having executable instructions encoded thereon, such that upon execution of the instructions by one or more processors, the one or more processors perform operations of: computing relative positions of multiple objects of interest; generating a feature representation by forming a matrix based on the relative positions; predicting movement of the multiple objects of interest by applying clustering to the feature representation and canonical correlation analysis; and controlling a device based on the predicted movement of the multiple objects of interest.
This invention relates to a system for predicting the movements of multiple objects of interest using computational techniques. The system addresses the challenge of accurately forecasting the trajectories of dynamic objects in real-world environments, which is critical for applications such as autonomous navigation, robotics, and surveillance. The solution involves a computer program product that processes relative positional data of the objects to generate a predictive model. The system computes the relative positions of the objects and constructs a feature representation in the form of a matrix based on these positions. This matrix is then analyzed using clustering techniques to group similar movement patterns and canonical correlation analysis to identify relationships between the objects' movements. The combined analysis produces a predictive model that forecasts the future positions and trajectories of the objects. The system then uses these predictions to control a device, such as an autonomous vehicle or robotic system, enabling it to respond to the anticipated movements of the objects in its environment. The approach improves decision-making in dynamic scenarios by leveraging statistical and machine learning methods to enhance movement prediction accuracy.
9. The computer program product of claim 8 , wherein the device includes a display.
A system and method for managing data processing tasks in a computing environment involves a device with a display that executes a computer program to perform specific operations. The device is configured to receive input data, process the data according to predefined rules, and generate output data. The processing includes analyzing the input data to identify relevant information, applying transformations to the data based on the identified information, and validating the transformed data to ensure accuracy. The system also includes a display for presenting the processed data to a user, allowing for real-time monitoring and interaction. The display may show visual representations of the data, such as graphs or tables, to facilitate user understanding. The device may further include input mechanisms, such as a keyboard or touchscreen, to allow users to provide additional data or adjust processing parameters. The system is designed to improve efficiency in data handling by automating repetitive tasks and reducing manual intervention. The display enhances usability by providing clear visual feedback, making it easier for users to interpret and act on the processed data. This approach is particularly useful in environments where timely and accurate data processing is critical, such as in financial analysis, scientific research, or industrial monitoring.
10. The computer program product of claim 8 , wherein the device includes a motor.
A system for controlling a device with a motor, where the device is configured to perform a task based on user input. The system includes a computer program product stored on a non-transitory computer-readable medium, containing instructions that, when executed by a processor, cause the device to receive user input specifying a task, analyze the task to determine a sequence of actions, and generate control signals to operate the motor to perform the sequence of actions. The motor is part of the device and is used to execute the task, such as moving components or adjusting settings. The system may also include a user interface for receiving input and a processor for executing the instructions. The device may be a robotic system, an automated machine, or another motor-driven apparatus designed to perform tasks based on user commands. The instructions further ensure the motor operates within safe parameters, such as speed and torque limits, to prevent damage or malfunction. The system may also include feedback mechanisms to monitor motor performance and adjust operations in real-time. The overall goal is to provide an automated solution for task execution using motor-driven devices, improving efficiency and precision in performing the specified tasks.
11. The computer program product of claim 8 , wherein the one or more processors further perform the operation of generating pairs of tactical feature vectors.
A system and method for generating tactical feature vectors in a computational environment involves processing data to extract and analyze tactical features for decision-making or pattern recognition. The system includes one or more processors configured to receive input data, such as sensor readings, user inputs, or environmental data, and process this data to identify relevant tactical features. These features are then encoded into feature vectors, which are numerical representations used for further analysis, machine learning, or real-time decision support. The system may also include a memory for storing the input data, feature vectors, and intermediate processing results. The processors generate pairs of tactical feature vectors by comparing or combining individual feature vectors to derive relationships, similarities, or differences between them. This pairing process may involve statistical analysis, machine learning techniques, or domain-specific algorithms to enhance the accuracy and relevance of the extracted features. The generated pairs can be used for tasks such as anomaly detection, predictive modeling, or tactical decision-making in applications like military operations, autonomous systems, or cybersecurity. The system ensures efficient processing and real-time responsiveness by optimizing the feature extraction and vector generation steps.
12. The computer program product of claim 11 , wherein the canonical correlation analysis is performed using the pairs of tactical feature vectors.
This software predicts the explicit movements of multiple objects, such as members of an adversary team. It begins by computing the relative positions of these objects. From these relative positions, it generates a feature representation by forming a matrix. The software then generates pairs of tactical feature vectors, which represent embedded tactical formations (e.g., for a home team and an adversary team). To predict movement, the system applies clustering to the feature representation and performs Canonical Correlation Analysis (CCA). This CCA is specifically executed using the generated pairs of tactical feature vectors to establish relationships between different sets of tactical formations and predict future movements. Finally, a device is controlled based on these predicted adversary movements. ERROR (embedding): Error: Failed to save embedding: Could not find the 'embedding' column of 'patent_claims' in the schema cache
13. The computer program product as set forth in claim 8 , wherein controlling the device includes causing a camera to orient based on the predicted movement.
A system and method for controlling a device, such as a camera, based on predicted movement. The technology addresses the challenge of dynamically adjusting device orientation to capture or track moving objects or scenes in real-time. The system predicts the movement of a target, such as a person or object, and adjusts the camera's orientation to maintain focus or alignment with the predicted path. This involves analyzing motion data, such as velocity and direction, to determine future positions and automatically adjusting the camera's angle or position accordingly. The system may use sensors, image processing, or machine learning to refine predictions and ensure accurate tracking. The solution improves automation in surveillance, photography, and robotics by reducing manual adjustments and enhancing responsiveness to dynamic environments. The invention ensures continuous and precise alignment with moving targets, improving efficiency and accuracy in applications requiring real-time tracking.
14. The computer program product as set forth in claim 8 , wherein the canonical correlation analysis (CCA) maximizes the following objective function: CCA comp = arg max u , w ∑ n = 1 N ( u T h n ) ( v n T w ) ∑ n = 1 N u T h n h n T u ∑ n = 1 N w T v n v n T w = arg max u , w u T C hv w u T C hh u w T C vv w wherein u and w are CCA components that project data onto a shared embedding and C hh , C vv , C hv are covariance matrices, the tactical formations of a home team and an adversary team are embedded into vectors h and v, respectively, N is the total number of tactical formations during a given time period, and the multiple objects of interest are the members of the adversary team.
This invention relates to a computer program product for analyzing tactical formations in team-based conflicts, such as military or sports scenarios, using canonical correlation analysis (CCA). The problem addressed is the need to identify and quantify relationships between the tactical formations of opposing teams to predict or understand strategic interactions. The solution involves embedding the formations of a home team and an adversary team into vectors, then applying CCA to project these vectors into a shared embedding space. The CCA maximizes an objective function that correlates the formations of both teams, optimizing the alignment of their tactical patterns. The objective function is defined as the ratio of the cross-covariance between the home team's formations (h) and the adversary team's formations (v) to the product of their individual covariance matrices (C_hh and C_vv). This approach allows for the identification of key relationships between the movements and positions of the adversary team members, enabling better strategic decision-making. The method leverages statistical correlation to uncover hidden dependencies in the formations, improving situational awareness and predictive capabilities.
15. A computer implemented method for predicting movements, the method comprising an act of: causing one or more processers to execute instructions encoded on a non-transitory computer-readable medium, such that upon execution, the one or more processors perform operations of: computing relative positions of multiple objects of interest; generating a feature representation by forming a matrix based on the relative positions; predicting movement of the multiple objects of interest by applying clustering to the feature representation and canonical correlation analysis; and controlling a device based on the predicted movement of the multiple objects of interest.
This invention relates to a computer-implemented method for predicting the movements of multiple objects of interest, such as in autonomous systems or robotic navigation. The method addresses the challenge of accurately forecasting object trajectories in dynamic environments where interactions between objects influence their paths. The system computes the relative positions of the objects to establish spatial relationships, then constructs a feature representation by organizing these positions into a matrix. This matrix captures the dynamic interactions between objects, enabling more accurate movement predictions. The method applies clustering techniques to group similar movement patterns and canonical correlation analysis to identify correlations between the objects' movements. Based on these predictions, the system controls a device, such as a robot or autonomous vehicle, to navigate or respond to the anticipated movements. The approach improves decision-making in real-time applications by leveraging statistical and machine learning techniques to enhance trajectory forecasting. The invention is particularly useful in scenarios requiring precise movement prediction, such as collision avoidance, path planning, or human-robot interaction.
16. The method of claim 15 , wherein the device includes a display.
A system and method for enhancing user interaction with a device through visual feedback. The device includes a display that provides real-time visual indicators to guide user actions, such as aligning or positioning the device relative to an object or environment. The display may show alignment markers, status indicators, or other visual cues to assist in proper usage. The device may also include sensors to detect environmental conditions or user inputs, which are processed to generate appropriate visual feedback on the display. This feedback helps users achieve precise positioning, calibration, or interaction with the device, improving accuracy and usability. The system may be applied in various fields, including augmented reality, robotics, or industrial applications, where visual guidance enhances operational efficiency and reduces errors. The display may be integrated into the device or connected wirelessly, and the visual feedback can be dynamically adjusted based on real-time data from the sensors. This approach ensures users receive clear, actionable information to optimize device performance and user experience.
17. The method of claim 15 , wherein the device includes a motor.
A system and method for controlling a device with a motor to perform a specific function. The device includes a motor that drives a mechanical component to achieve a desired operation. The motor is controlled by a controller that receives input signals from sensors or user interfaces to adjust the motor's operation. The controller may regulate motor speed, torque, or direction based on predefined parameters or real-time feedback. The system ensures precise and efficient operation of the device by dynamically adjusting motor performance to meet operational requirements. The motor may be an electric, hydraulic, or pneumatic type, depending on the application. The device can be used in various industries, including robotics, automation, and industrial machinery, to enhance productivity and accuracy. The method optimizes motor control to improve energy efficiency, reduce wear, and extend the lifespan of the device. The system may also include safety features to prevent overheating, overloading, or other operational hazards. The motor control method ensures reliable and consistent performance under varying load conditions.
18. The method of claim 15 , wherein the one or more processors further perform the operation of generating pairs of tactical feature vectors.
This invention relates to a method for processing tactical data, particularly in military or defense applications where real-time analysis of battlefield conditions is critical. The method addresses the challenge of efficiently extracting and comparing tactical features from sensor data to improve situational awareness and decision-making. The system uses one or more processors to analyze incoming data, such as radar, satellite, or other sensor inputs, to identify relevant tactical features like enemy positions, friendly forces, or environmental obstacles. These features are then converted into numerical vectors, which represent the key characteristics of each feature in a structured format. The method further includes generating pairs of these tactical feature vectors to enable comparative analysis, allowing for the assessment of relationships, distances, or similarities between different features. This pairwise comparison helps in identifying patterns, threats, or opportunities that may not be apparent from individual feature analysis alone. The method may also involve filtering or prioritizing features based on their relevance or urgency, ensuring that critical information is highlighted for decision-makers. By automating the extraction and comparison of tactical features, the system enhances the speed and accuracy of battlefield assessments, supporting more effective command and control operations.
19. The method of claim 18 , wherein the canonical correlation analysis is performed using the pairs of tactical feature vectors.
This invention relates to a method for analyzing tactical feature vectors using canonical correlation analysis (CCA) to improve decision-making in dynamic environments, such as military operations or autonomous systems. The method addresses the challenge of extracting meaningful relationships between multiple sets of tactical features, which are often high-dimensional and noisy, to enhance situational awareness and predictive accuracy. The method involves generating pairs of tactical feature vectors from different data sources or time steps, where each vector represents a set of relevant attributes such as sensor readings, positional data, or behavioral patterns. These pairs are then processed using canonical correlation analysis to identify latent correlations between the vectors, revealing underlying dependencies that may not be apparent through traditional statistical methods. The CCA step computes canonical variates, which are linear combinations of the original features that maximize correlation between the paired vectors, thereby simplifying the data while preserving critical relationships. The method further includes refining the canonical variates to improve robustness against noise and outliers, ensuring that the derived correlations are reliable for real-time applications. The output of the analysis can be used to optimize decision-making processes, such as threat assessment, resource allocation, or adaptive control in autonomous systems. By leveraging CCA on tactical feature vector pairs, the method provides a more accurate and efficient way to model complex interactions in dynamic environments.
20. The method as set forth in claim 15 , wherein controlling the device includes causing a camera to orient based on the predicted movement.
A system and method for controlling a device, such as a camera, based on predicted movement. The technology addresses the challenge of maintaining accurate tracking or imaging in dynamic environments where movement of the device or its target is unpredictable. The method involves analyzing sensor data, such as from inertial measurement units (IMUs) or other motion sensors, to predict future movement of the device or an object within its field of view. Based on this prediction, the device is controlled to adjust its orientation, position, or other operational parameters to compensate for the anticipated movement. For a camera, this may involve adjusting its angle, zoom, or focus to ensure continuous tracking of a moving target. The system may also incorporate feedback loops to refine predictions and adjustments in real time. The invention improves stability, accuracy, and responsiveness in applications such as surveillance, robotics, autonomous vehicles, and augmented reality, where precise tracking of moving objects is critical. The method ensures that the device remains aligned with the target despite external disturbances or unpredictable motion.
21. The method as set forth in claim 15 , wherein the canonical correlation analysis (CCA) maximizes the following objective function: CCA comp = arg max u , w ∑ n = 1 N ( u T h n ) ( v n T w ) ∑ n = 1 N u T h n h n T u ∑ n = 1 N w T v n v n T w = arg max u , w u T C hv w u T C hh u w T C vv w wherein u and w are CCA components that project data onto a shared embedding and C hh , C vv , C hv are covariance matrices, the tactical formations of a home team and an adversary team are embedded into vectors h and v, respectively, N is the total number of tactical formations during a given time period, and the multiple objects of interest are the members of the adversary team.
This invention relates to analyzing tactical formations in competitive scenarios, such as sports or military engagements, to identify patterns and correlations between the movements of a home team and an adversary team. The problem addressed is the need to extract meaningful insights from complex, multi-object interactions in dynamic environments where the behavior of one team influences the other. The method uses canonical correlation analysis (CCA) to maximize an objective function that quantifies the relationship between the tactical formations of the home team and the adversary team. The CCA components, denoted as u and w, project the data into a shared embedding space, allowing for the identification of correlated patterns. The tactical formations of the home team are represented as vectors h, while those of the adversary team are represented as vectors v. The analysis involves computing covariance matrices C_hh, C_vv, and C_hv, which capture the relationships between the formations of the two teams. The objective function is maximized to find the optimal projections u and w, which reveal the strongest correlations between the teams' movements. This approach enables the identification of strategic interactions and predictive insights based on historical or real-time formation data. The method is particularly useful in scenarios where understanding adversarial behavior is critical, such as sports analytics or military strategy.
Unknown
March 10, 2020
Browse 5M+ US patents with plain-English claim translations and AI-generated analysis.